---
license: mit
pipeline_tag: text-generation
language:
- ja
tags:
- japanese
- qwen2
- mlx
- mlx-my-repo
inference: false
base_model: cyberagent/DeepSeek-R1-Distill-Qwen-14B-Japanese
---

# melt-adzuki/DeepSeek-R1-Distill-Qwen-14B-Japanese-Q4-mlx

The Model [melt-adzuki/DeepSeek-R1-Distill-Qwen-14B-Japanese-Q4-mlx](https://huggingface.co/melt-adzuki/DeepSeek-R1-Distill-Qwen-14B-Japanese-Q4-mlx) was converted to MLX format from [cyberagent/DeepSeek-R1-Distill-Qwen-14B-Japanese](https://huggingface.co/cyberagent/DeepSeek-R1-Distill-Qwen-14B-Japanese) using mlx-lm version **0.20.5**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("melt-adzuki/DeepSeek-R1-Distill-Qwen-14B-Japanese-Q4-mlx")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```